Red Balloon, Green Balloon, Sensors in the Sky - Eric Paulos

Red Balloon, Green Balloon, Sensors in the Sky - Eric Paulos

Red Balloon, Green Balloon, Sensors in the Sky

Stacey Kuznetsov 1 , George Noel Davis 2 , Eric Paulos 1 , Mark D. Gross 3 , Jian Chiu Cheung 2

Human-Computer Interaction Institute 1

Carnegie Mellon University

Pittsburgh, PA, USA

{stace, paulos}

Department of Electrical and

Computer Engineering 2

Carnegie Mellon University

Pittsburgh, PA, USA

{gnd, jccheung}

Computational Design Lab 3

School of Architecture

Carnegie Mellon University

Pittsburgh, PA, USA


Spectacle computing is a novel strategy for vibrantly

projecting information into the public sphere using

expressive and tangible media. We demonstrate an example

of this computing meme with large, glowing balloons that

change color based on input from attached air quality

sensors (exhaust, diesel, or volatile organic compounds). In

two public installations (city street and public park) and a

deployment with six everyday citizens, we invited

stakeholders to playfully explore and actively participate in

visualizing surrounding air quality. We also created a do-ityourself

(DIY) kit that includes a printed circuit board,

electronic parts and instructions for building the air quality

balloons. In a workshop, six non-expert users successfully

assembled functional balloons, validating our technology as

a DIY tool for public air quality visualization. Our

deployments and workshop highlight play and spectacle as

essential elements for public participation and activism. We

outline design guidelines for future spectacle computing

projects that engage stakeholders with environmental data

and empower them to transform urban landscapes.

Author Keywords

Spectacle computing, participatory sensing, air quality

ACM Classification Keywords

H5.m. Information interfaces and presentation (e.g., HCI):


General Terms



Low cost sensors, DIY (do-it-yourself) methods and

bottom-up initiatives have expanded environmental sensing

beyond the domain of scientists and experts, enabling

ordinary citizens to measure, analyze, and share

information about their environment. An emerging body of

HCI and ubicomp research has explored this mode of

distributed data collection by non-experts, referred to as

participatory sensing [8] or citizen science [39]. Recently,

Permission to make digital or hard copies of all or part of this work for

personal or classroom use is granted without fee provided that copies are

not made or distributed for profit or commercial advantage and that copies

bear this notice and the full citation on the first page. To copy otherwise,

or republish, to post on servers or to redistribute to lists, requires prior

specific permission and/or a fee.

UbiComp ‘11, September 17–21, 2011, Beijing, China.

Copyright 2011 ACM 978-1-4503-0630-0/11/09...$10.00.

Figure 1. Air quality balloons in public park (top left), diesel

balloon inflated (top right), DIY diesel kit fully assembled

(bottom left), and a VOC balloon close-up (bottom right).

citizen science projects have focused on handheld devices

such as cellphones and data loggers [e.g., 11] to collect

environmental data, and visualizations such as graphs,

charts, and maps [e.g., 16] to present this data to

communities of users. We introduce spectacle computing as

a playful and powerful approach to enable non-experts to

view environmental data and project this information into

the public sphere.

We developed large, glowing, air-quality sensing balloons

as a vehicle for stakeholders to explore and reflect on urban

air quality (Figure 1). The balloons change color in

response to one of exhaust gas, diesel, or VOC’s (volatile

organic compounds). These three pollutants continue to

exacerbate urban air quality problems, especially in the city

where we conducted this research, where air quality has

been rated as among the worst in the United States [4].

Diesel exhaust consists of fine particulate matter emitted by

engines and industrial processes [44]; exhaust gas—a

mixture of carbon monoxide, nitrogen dioxide, and

hydrocarbons—is produced by gasoline engines [9]; and

VOC’s (volatile organic compounds) originate from paints,

pesticides, or certain types of fuels [45]. Exposure to any of

these substances can lead to chronic respiratory illnesses,

not to mention the pollutants’ role in global climate change.

Research objectives

The following section describes social phenomena that

inspired spectacle computing and reviews related ubicomp

and HCI literature. We then detail the design of our air

quality balloons and present findings from two public

installations and deployments with six everyday citizens.

Finally, we introduce our DIY kit, which we distributed in a

workshop where six non-expert users successfully

assembled functional air quality sensing balloons. Our

findings offer insights into questions such as: 1) how do

playful spectacles influence public perceptions of air

quality; 2) how do everyday citizens use tangible media to

broadcast their concerns about air quality; and 3) what are

the implications of tangible media as a supporting material

for DIY making and grassroots activism


Artists have a long history of integrating “the spectacle”

into their work: from Allan Kaprow’s Happenings [27] and

the writings of Guy Debord and the Situationists in the

1960’s [14], to more contemporary tactical media artists

such as The Yes Men, Critical Art Ensemble, RTMark,

Preemptive Media, and Institute for Applied Autonomy.

The Situationists differentiated between passive subjects—

consumers of the spectacle—and those who transform their

own ideas, concerns, and passions into the spectacle itself.

This movement applied concepts of commodity fetishism

[33] to contemporary mass media to expose the common

politics of its day. Spectacle computing intentionally and

overtly foregrounds these ideas, using expressive

technologies to inspire new thinking, curiosity and beliefs.

Stakeholders who otherwise may not be aware of or care

about an issue are drawn into the spectacle.

Contrary to contemporary rhetoric of “invisible” interfaces

and seamless computing, we argue for a complementary

strategy explicitly designed to generate spectacles. First and

foremost, spectacles are difficult to ignore. The barrier to

engagement is thereby effectively lowered because

individuals need not download an application or carry

specific hardware. The spectacle is intentionally designed to

distract the individual or group’s attention. Moreover, it

invites people to engage in otherwise socially unacceptable

behaviors such as overt public voyeurism, gossip and

curiosity. Finally, it presents an acceptable context for

individuals to participate in the spectacle, in the spirit of a

Happening, even if such participation involves odd,

unusual, or socially awkward activities (i.e. willingly taking

and carrying around a glowing balloon).

We view these insights as experience design opportunities.

We observe how spectacle computing moves people from a

personal and private context, through public voyeurism, and

into readily carrying balloons around the city, thereby

creating their own spectacle. To be clear, spectacle

computing is not intended to mimic the experience of

yelling “fire” in a crowded theater, but to more tactfully and

expressively engage public audiences in issues of personal

or societal concern. While this approach is tangentially

related to FlashMobs, which draw large groups of people to

suddenly assemble and perform unusual acts in public

places, the goal of spectacle computing is to foster

discourse between stakeholders, technology and space

through the use of dynamic computing elements. Also

unlike FlashMobs, which may create a feeling of inclusion

and exclusion, spectacle computing invites open

participation from everyone.


We use balloons as an expressive medium for creating air

quality spectacles. Balloons are inherently playful: they

remind us of birthday parties, water balloon fights, street

fairs, weddings, or carnivals. Balloons are also functional:

for decades they have enabled scientific endeavors, ranging

from the NASA superpressure balloon which successfully

flew over Antarctica [34] to the National Weather Service

radiosondes (sensor packages) deployed on weather

balloons [36], as well as a milieu of grassroots aerial

photography projects [e.g., 19]. Lastly, balloons are

compelling and visual: they have shaped artistic and

political expressions, including the popular song 99

Luftballons [35], Lamorisse’s film The Red Balloon [32],

the literary work More Sky [41], and most recently,

interactive balloon installations such as Open Burble [20].

This vast range of artistic and functional balloon projects

inspired us to utilize large glowing weather balloons as a

playful expression of a serious concern—public air quality.


Ubicomp and HCI communities have explored a range of

participatory sensing platforms to enable citizens to monitor

their environment. Examples include handheld sensing

devices [10] cellphones [23] sensors installed in private

homes [28], or sensors deployed on moving vehicles [3],

bicycles [16] and pigeons [12]. Most projects employ

conventional visualization methods: line graphs [28], maps

[11] or numeric data [16]. More expressive and public

visualizations include WearAir—a T-shirt that responds to

air quality with interactive lights [29], CO2RSET—a corset

that constricts when the air becomes toxic [37], the

pollution e-sign, which hijacks passersby mobile phones

with air quality data [24], and Air de Paris Balloon—a

giant balloon tethered over the city of Paris to show data

collected from several sensing stations [1]. Unlike prior

work, our balloon visualization enables stakeholders to

transform and produce urban lightscapes with sensor data.

Urban lightscapes

Public displays and installations such as BlueTone [13],

Aarhus by Light [10], Amphibious Architecture [2], and

many others have enabled stakeholders to alter urban

lightscapes through sound, body movement or text

messages. Modular technologies including LED Throwie’s

[18], Light Bodies [43], and WallBots [31] allow

individuals (ranging from ordinary passersby to artists and

activists) to move and place interactive elements in urban

spaces. Drawing from this prior work, we created modular

sensors and reactive lights carried by weather balloons. Our

project is inspired by other creative balloon installations:

Burble- a floating cloud of balloons that is controlled by

moving a handlebar [20], and Wire and Balloon

Installation, which responds to touch with sound [40], to

name a few.

Support tools for DIY practices

A range of HCI initiatives have focused on technologies

that involve communities of non-experts in hands-on

making. Examples include Sketching in Hardware—a

creative workshop for physical prototyping [22]; Simple

Haptics—a suite of tools for creating haptic interfaces; or

numerous e-textile workshops with the LilyPad as a

ubiquitous tool for education [e.g., 5, 6]. In addition, prior

work has explored the values and practices of DIY

communities through conference workshops [e.g., 7, 26],

surveys [30], or qualitative fieldwork [42]. We contribute to

this research by introducing a low-cost DIY kit that enables

non-experts to assemble their own interactive air quality

balloons. We evaluate our approach by conducting a DIY

balloon-making workshop with hobbyists from a local DIY



Our circuit is powered by a rechargeable lithium polymer

battery and relies on one of two inexpensive sensors from

Figaro: a VOC sensor or a dual function Diesel/Exhaust

sensor [17]. The latter measures either diesel or exhaust on

separate balloons, and each balloon is labeled with the

corresponding pollutant (‘voc’, ‘diesel’, or ‘exhaust’).

PICAXE [38], a low-cost ($1.50) microcontroller,

processes sensor data and controls a tri-colored LED. This

LED is inserted into the balloon; the sensor,

microcontroller, and battery are mounted outside using wire

and electrical tape. Balloons are illuminated based on

surrounding air quality, glowing green, yellow, or red to

indicate ‘low’, ‘average’, or ‘high’ pollution levels.

Balloons are inflated using helium and their latex material

naturally diffuses LED light, resulting in a large, evenly

‘glowing’, floating orb.

Ambient light thresholds

To support Do-It-Yourself (DIY) principles of easy access

to materials and methods for creation by non-experts, we

constrained the design of our system by choosing low-cost

and low-fidelity sensors. We empirically determined our

illumination thresholds (sensor values for green, yellow, or

red balloon lighting) by gathering sensor data around

Pittsburgh over the course of one week. We visited five

different neighborhoods and collected readings indoors,

outdoors, and at varying distances from busy streets, as well

as in (traffic-free) parks. We used the observed variance in

sensor output to determine thresholds below, within, and

above the average sensor output for green, yellow, and red

lighting respectively. Consequently, our balloons show air

quality relative to the data collected in our city. Our display

choice appropriately affords open interpretation through

relative threshold lighting than exact numbers.


We deployed about thirty balloons throughout the city of

Pittsburgh, Pennsylvania, USA over the course of one

night. The authors tethered balloons around two different

locations (in a public park and along a city street),

observing public interactions with the installation. In

addition, six stakeholders were invited to participate in our

project, placing and photographing air quality balloons in

public spaces throughout the city.


We selected two locations to install our air quality balloons:

a public park and a city street (Figure 2), expecting to see a

range of balloon colors due to different amounts of traffic,

people, etc. in the two spaces. Our installation coincided

with extreme humidity (over 80%) and a PM2.5 (fine

particulate matter) alert, causing many of the balloons to

turn red. We inflated the balloons on site, tethering them to

trees and benches (in public park) and on parking meters,

cars and street signs along a city block. In addition to

observing public interactions with our project we

interviewed individuals who approached us and invited

them to take some balloons. The installation was after dusk

(around 9.30pm), so we encountered few people in the

public park but numerous passersby approached us along

the city block. We spent about one hour in each location.

Public reactions

Consistent with our framing of spectacle computing, nearly

everyone who saw our installation was compelled to stop or

slow down. In the public park and along the city block

people expressed curiosity about the large, mysteriously

glowing balloons. Most initial reactions suggested awe at

the size and illumination: “I thought there was a party

down the street”, “My kids love your balloons”, “Awesome

balloons”, etc. Most people understood the significance of

the red-yellow-green lighting upon closer inspection—

when they read the balloon labels, or after they asked us to

confirm that the balloons ‘showed air quality’. People

reacted positively: “Keep it up, I dig it!”, “Diesel exhaustthat’s

a very bad thing” or, “Awesome, so how can we

make money with this” Not surprisingly, individuals who

Figure 2. Air quality balloon installations in public park (top

left & right); and city street (bottom left & right).

Figure 3. Interview setup: participant drawing on map

transparency prior to sorting air quality factors (cards).

lived in the neighborhoods . where we placed our

installations related more intimately, with one person


This is my neighborhood and I know everybody, so I

would probably put them [balloons] wherever I

wanted. I think you should have more people with


Several people asked us if they could take a few balloons.

One person excitedly pointed out:

I think anything that’s big and shiny will get America’s

attention… I know a little bit about VOCs so I can

spread the word… Word, I feel powerful now, can she

[friend] have one

Thus, when first noticing the installation, passersby reacted

with excitement and curiosity. Upon further reflection,

however, stakeholders began to more critically discuss the

visualization, the environment, and their neighborhood.

Some further speculated on the role they can play in

‘spreading the word’.


In addition to our public installations, we recruited six

individuals by advertising our project on Craigslist (a

general-purpose classified advertising website) and local

biking forums. We invited participants for a semi-structured

interview exploring their relationships with urban spaces.

We asked them to mark locations on a map of Pittsburgh

associated with words such as “work”, “play”, “unhealthy”,

“alienating”, “personal”, “boring”, “interesting”, etc. In

addition, participants identified places where they would

like to monitor and broadcast air quality (Figures 3 & 4).

Lastly, participants completed a card sorting exercise,

ordering air quality factors such as traffic, household

cleaning products, pollen, pesticides, industrial byproducts,

etc., from most to least harmful for their health.

Upon completing the interview, we showed participants the

air quality balloons and offered them up to three (choosing

between VOC, diesel, exhaust). We encouraged participants

to photograph and use the balloons as they saw fit during

the course of the night: to carry them all night, leave them

temporarily or permanently in a place, etc. All participants

returned for a follow-up interview to share their pictures of,

and experiences with, the air quality balloons. We recorded

interview audio, and reference data owing to individual

study participants as P1-6.

Figure 4. A subset of transparency maps drawn by a participant.


Stakeholders and spaces

Our pre-balloon interviews reveal a wide range of

participants’ concerns, values, and relationships with public

spaces in Pittsburgh. P1 and P2 (a couple) are graduate

students who spend time outdoors biking, running and

playing sports such as rugby. P3 is working on a medical

residency, commuting regularly by bicycle along bike trails,

parks, and city streets. P4 and P5 are artists (also a couple),

and P6 is a student; all three traverse the city by bus. None

view themselves as political or environmental activists, and

only one sporadically attends social eco-themed events. The

six participants occupy very different neighborhoods,

working and living in four distinct regions of Pittsburgh.

Participants also have divergent perceptions of urban

spaces: for instance, P1 and P2 hang out in an ‘interesting’

neighborhood that P3 and P4 consider ‘alienating’.

Despite these differences, all participants frequent the same

public parks and expect air quality to be cleaner there. They

also consider the same neighborhoods to be unhealthy

spaces, describing them as ‘rundown’, ‘industrial’, or

having ‘construction’, ‘traffic’, or ‘bad smell’. Moreover,

everyone identifies vehicle traffic as the most harmful

factor for health in terms of air quality. Pesticides, trash,

dust, and pollen are considered least harmful because

participants are less frequently exposed to these factors.

Sensing and broadcasting air quality

When asked to identify places where they would sense air

quality, all participants mentioned sites they considered

dirty, polluted, or unhealthy. In addition, several people

wanted to monitor places they felt were clean (“I was

thinking that- probably is clean there… At [park] would be

more positive”, P1). All participants wanted to broadcast air

quality data to the general public, for instance via a ‘blimp’

(P4), or atop a bridge (P6), cultural district (P3), or

downtown (P1& P2). Many also suggested locations that

target specific audiences. For instance, P1 and P2 suggested

spaces most visible to other students:

These two [pointing on map] would be college

students, because that would be the target

demographic for people who’d want to change it. (P1)

Meanwhile, P3 suggested an ‘eco-friendly’ neighborhood:

P3: Lawrenceville, because it’s full of eco-people that

might be a little more aware of it. Like a little more

open and receptive towards it, so people that are more

active in it and might be more receptive to the


I: More receptive how What are you thinking

P3: People that are more willing to change behaviorso

to do things about it or to tell other people about it.

Balloons in the hands of urban stakeholders

When offered the three types of balloons, five of the six

participants wanted to measure VOC. P3 explained:

That’s something that you can’t get an idea of just

visually… ‘cause you know, when you’re sucking in

car exhaust—at least high levels, you notice that

you’re sucking in car exhaust. But with organic

compounds, it is what it is—like you don’t know. (P3)

The two couples (P1&P2 and P4&P5) and P6 also took an

exhaust or diesel balloon. Participants brought balloons to

different parts of the city: public park and downtown city

center (P4&P5), to her home and later to a party (P3), along

a city street to a neighborhood where participants tend to

hang out (P1&P2), and around a college campus (P6). All

six participants made an effort to visit a part of the city

outside of their evening plans because of the balloons.


Participants reported that their balloons changed colors

based on location and activity, with most people finding

these changes ‘interesting’ or ‘surprising’. Participants

tended to rationalize balloon behavior: for instance, P1 and

P2 noticed that the diesel balloon turned red whenever they

put it inside their house, proposing several explanations:

P1: We came up with theories but none of them that…

not necessarily impossible, but none of them even

borderline likely.

I: A-ha, like what

P1: Like the stove- the gas that we use for the stove

could be producing diesel... but that doesn’t make any

sense. The other theory was that there’s a lot of

humidity in the house- like much more than outside,

and the humidity… affects a lot of things.

Likewise, P3 told us that her VOC balloon reacted to doors

opening and closing in her house:

Later that night, P3 and her friends noticed the balloon

turning red outside of a bar. P3 suggested that this was

caused by chemicals from a nearby children’s hospital.

Public attention: wanted

All six participants intentionally brought the balloons to

places where they would encounter more people, reflecting

positively on public attention. P2 told us:

I think they [public reactions] were pretty positive,

everybody seemed pretty excited about the balloons.

Similarly, P4 noted people’s immediate reaction to red

balloon colors:

It caught a lot of attention, it got people almost instant

awareness, especially if it was red at the time they saw

it: they’re like Wow! That’s… concerning!!

All participants interacted with strangers, asking them to

help photograph the balloons and explaining the project to

passersby. Consequently, observers (passersby) began to

reflect on other air quality issues in their lives. For instance:

P1: We got pictures with people, we were like ‘here

can you hold these’ and we explained what we were

doing to them’.. and then [laughing] towards the end

of the night people had progressively more interest in

it. We had one group ask us like in detail what we were

doing and we talked to them for like 5-10 minutes.

I: And what did they say when you told them

P1: They were like oh, that’s interesting… and then

one guy explained how the levels in his hometown were

above what they’re supposed to be because of the


Moreover, balloons inspired discussion about other

expressions of air quality. P5 detailed her experience with a

passerby who suggested logging and storing the data:

I left it in my house for a couple of hours and the

interesting thing was- it was red when my door was

closed and green when my door was open. (P3)

P3 noted that perhaps VOC’s tend to accumulate in her

house due to poor ventilation:

It does make me wonder if my house worse because it’s

not like aerated or ventilated. (P3)

P3’s friend also pointed out that the VOC balloon reacted to

hair spray:

I was going to a costume party, and I was a giraffe so I

had lots of hair spray going on… and then I like- stood

near it and it suddenly went like really red.

Figure 5. Air quality balloons placed by study participants:

diesel balloon downtown (top left), VOC balloon near public

hospital (top right), VOC & diesel balloons in an underpass

(bottom left), and VOC balloon in public park (bottom right).

P5: One guy recommended having… an actual digital

log of how it changes color, and then just having

certain paths all ready that you go down. And then you

could record the changes over a period of time…

P5: …and then that would really track how things

change, because I mean, just walking form one place

to another, we were seeing changes left and right.


We asked participants what they would do if they could use

our balloons for longer periods of time. One person (P3)

wanted to place balloons all around a children’s hospital to

monitor VOC levels, and another suggested to permanently

insteall balloons at street intersections and in public parks.

Participants were skeptical about reporting this data to local

authorities, with one person explaining:

I guess we usually say that we’d want to show it to the

mayor, except I don’t really believe in it. (P5)

Instead, people wanted to broadcast the data they collected

to as many people as possible:

If you show it to just one person who’s either back

higher [in the government]… would just toss it away

but if you get more- dozens, hundreds, thousands,

millions of people, then you can get a movement. You

need a ton of people to start a revolution, you can’t

just have one person start a revolution. (P6)

Ultimately, participants wanted to have more balloons for

longer periods of time in order to collect and share data

with the general public. Most participants also suggested

showing data on a map, with P3 emphasizing visual (nonnumeric)


I don’t wanna read things like- this is however many

parts per million… I just want to say like- this is bad,

this is better, this needs to be better because that’s all

you really wanna boil it down to. (P3)

In addition, participants wanted to show data specifically to

people whose behavior might be causing poor air quality:

We pretty much rely on either public transportation or

a mode of walking or biking or whatever, there’s really

not much we can do to make less exhaust or volatile

chemicals individually… But like I said, we can spread

it to people who we know that may be in cars. (P4)

To summarize, our deployment with six diverse individuals

inspired participants to explore new (non-routine) spaces

with the balloons, as well as seek public attention and

discuss the project with strangers. Reflecting on the study,

participants expressed interest in using balloons for longer

periods of time and in showing the visualization to other

stakeholders such as eco-friendly neighborhoods or those

who might be causing poor air quality.


Lastly, we created a DIY kit to enable non-experts to

assemble functional air quality balloons. Here, we describe

Figure 6. PCB board for the DIY balloon kit (left) and

instruction booklet for assembling the board (right).

our kit and circuit board, . and a workshop where six

participants assembled their own working balloons.

DIY balloon kit

Our DIY balloon kit includes a custom circuit board that

can be populated with a VOC or exhaust/diesel sensor. For

the dual-function exhaust/diesel sensor, one of two pairs of

solder pads must ‘shorted’ (soldered across) to supply input

from either the exhaust or the diesel pin of the sensor. The

board is powered by a regular 9 volt battery or two

rechargeable 3.7 volt lithium ion polymer (LiPo) batteries.

The circuit relies on two PICAXE-8M chips, with each

modulating the pulse width of a red or a green LED channel

based on input from the sensor. The PCB also supports an

optional piezo vibration sensor in order to alter balloon

brightness based on movement (e.g., tug on the balloon

string). Similar to our earlier design, a tri-colored LED

resides inside the weather balloon, connected to the PCB,

which hangs below the balloon neck. The board is

intentionally designed for a non-expert to assemble, with

ample spacing between through-hole solder pads and all

components clearly labeled (Figure 6, left). In addition, we

created a booklet with step-by-step picture instructions for

assembling the board (Figure 6, right). Each DIY kit costs

under $25 (VOC) or $35 (exhaust/diesel) and includes the

PCB, instruction booklet and all components (including an

air quality sensor, weather balloon, tri-colored LED, piezo,

two pre-programmed PICAXE chips, battery, and resistors).

DIY balloon workshop

Working with a local DIY community in Pittsburgh, we

conducted a free air quality balloon-making workshop with

six participants (Figure 7). The workshop began with brief

discussion about attendees’ backgrounds and motivations

for participating. Each participant then chose one of VOC,

exhaust or diesel balloon kits. Participants worked

independently to assemble kits by following the included

instructions, with most completing a board in about 30

minutes. Workshop organizers then assisted participants in

inflating balloons and attaching batteries to the PCB (with

tape). A few days later, an informal follow-up interview

was conducted by phone or email to gather participants’

feedback. Participants were compensated $15 for

completing this interview. We recorded audio and

photographed the workshop; we reference data owing to

individual workshop participants as W1-6.


Our workshop was hosted by a local community that

sponsors projects ranging from: creating hacky sack

footbags by filling balloons with sand; to working with

electroluminescent (EL) wire; to “yarn bombing and trying

to incorporate social feedback and small bits of electronics

into it” (W6); or “high altitude weather balloons to take

pictures of the curvature of the earth” (W1). Workshop

participants (ages mid 20’s to late 30’s, 5 male, 1 female)

have been involved with the DIY organization for various

amounts of time: more than 2 years (W1); over a year (W2,

W3, W6); 6 months (W4), and just under a month (W5).

Participants’ backgrounds varied, including degrees and

work in design, engineering and English. Participants had

some familiarity with soldering and different degrees of

proficiencies with electronics—from relative beginner

(“other kits I've assembled were much simpler”, W6; “my

first introduction to soldering was the class that [DIY

community] had in like September [6 months ago]”, W4),

to working on electronics projects on a weekly basis over

the past few years (W1, W2, W3). None of the participants

had previously worked with the PICAXE chip or Figaro air

quality sensors. One participant owns a similar carbon

monoxide sensor, but had not yet used it in a project.


All participants indicated that they were drawn to the

workshop for personal enjoyment: “a project that integrates

balloons with technology is just kinda fun for us” W1; “I

just thought it sounds like a really neat project” W3.

Although participants were not environmental activists,

some have previously thought about air quality in

Pittsburgh. W5 was concerned with air and soil near his


I always kinda wonder about the soil- you know, like

the lead… I wanna put this [balloon] at [work place]

so people can see it as they drive over the [bridge

name] bridge because it’s got a lot of visibility. (W5)

Similarly, W2 noted, “I’ve just come to accept that

Pittsburgh air quality sucks”, while W3 was interested in

sharing the project with his daughter:

I have a 3-year old and I’m really curious if it clicks…

the concept of air pollution doesn’t really mean much

to her but I’m curious to see if see if with the colors it

kinda clicks. (W4)

Assembling the DIY balloon kit

Most participants fully assembled their board in about 30

minutes, with only one participant taking over an hour due

to a malfunctioning transistor. Throughout the workshop,

participants asked several questions about the hardware

(“we’ll want more information about the kit—just even like

the components or the source code… just because it’s kinda

who we are, we like to explore things.” W5,), as well as a

few clarification questions about the instructions (e.g., why

are the steps in a certain order; or why/how to solder across

the pads for the diesel/exhaust sensors). Participants were

excited to turn on their balloons for the first time (“Testing

the balloon! I felt like six [years old] all over again”, W3;

“It was really fun to see the balloon light up for the first

time, to see them all together in the dark at the shop”, W6)

W4 also noted that unlike kits, which require programming

after assembly, the balloon kit worked immediately:

The neat thing about this kit is you come out with a

functional product. We were able to play with it

immediately. It wasn’t like… here’s the product now go

write the software that does something cool. (W4)

Participants’ experiences with balloons

After assembling the kits, participants were encouraged to

interact with their balloons throughout the night as they saw

fit. Several participants took the balloons to a restaurant,

but “it was really late and we couldn’t figure out where we

could take them where people would be” (W5). W4 brought

his balloon home to show his daughter the following day:

She and I spent some time talking about it, I tried to

explain it to her… I did the exhaust, so we did talk a

little bit about the smoke that comes out of cars (W4)

W3 tried to use long exposure photographs to capture his

balloon changing color as he walked from a street

intersection into the park (Figure 7, bottom right):

I walked from the intersection with traffic to the

playground in the park to see how the colors would

change. Also I figured it would make some cool

pictures with long exposure… (W3)

W2 discussed the project with his friends in India:

All of us agreed that it would be a great idea to have

this at several locations thorough-out Bangalore. The

air quality is noticeably different in different parts of

the city and the weather forecasts give pollution

readings only from a few places. (W2)

Likewise, W5 hoped to initiate dialogue by sharing his

Figure 7. Workshop participant assembling a DIY balloon kit

(top left); kit fully assembled (top right); fully-functional

balloons built by workshop participants (bottom left);

participants’ long exposure photo of balloon (bottom right).

images (“I’m hoping I can get some people responding on

my Flickr pool… I think people would ask me what it’s

about”, W5). W6, whose balloon deflated by the following

afternoon speculated on using it for a whole day, “to see

how and if it changed in the various places, and it would be

fun to see people's reactions to it as well”. Lastly,

participants had numerous ideas for modifying the kit, from

adding data logging (“so you could look back and see how

things changed”, W6), to incorporating the kit into a “semipermanent

system using plastic globes (lawn lights) on

pipes” (W2). In summary, our workshop enabled six

participants to assemble their own functional balloons, and

interactions with the project led to exploration and

speculation on expanding the project to include other

functionality or be deployed in different regions.


We presented the design of our air quality sensing circuit,

which was mounted on weather balloons and distributed to

stakeholders to visualize concentrations of VOC’s, exhaust

and diesel around our city. We detailed three deployments

of this technology: 1) public installations in city street and

park; 2) deployment with six urban stakeholders; and 3) a

workshop where six DIY hobbyists who were previously

unfamiliar with the project and its underlying technology

assembled functional balloons.


We intentionally chose low-cost sensors and pursued DIY

methods in order to position our technology as a tool for

bottom-up movements by ordinary citizens. Admittedly,

this approach does not achieve precise air quality

measurements, and our work faces the limitations of inexact

and un-calibrated sensors. Our visualization appropriately

represents sensor readings by illuminating balloons green,

yellow, or red based on whether the sensors read below, at,

or above average values that were collected throughout the

city. We believe there is tremendous value in

communicating even these relative measurements of air

quality to the public (as noted by one participant who did

not want to read more detailed ‘parts per million’ data).


We conclude with three implication areas that emerged

from our work, suggesting opportunities for future research

in this exciting domain of spectacle computing.

The importance of play

Our work inspired participants who were not environmental

activists to assemble and interact with air quality balloons

by incorporating elements of play.

Playful DIY making

Our balloon kit appealed to a local DIY community as a

project that is “just kinda fun for us”. The simple, clearlylabeled

PCB and step-by-step image instructions enabled

non-expert users to create what they considered to be

“functional products”, emphasizing the “fun” of “seeing the

balloon light up” or feeling “like 6 [years old] again”. Our

findings suggest involving users in playful DIY making as

an approach for inspiring environmental exploration.

Future systems can support easy assembly and modification

of the underlying technology to prompt ideas for creatively

augmenting the project or speculations about broader issues

such as air quality, as we saw in our workshop. New, even

more flexible, easy-soldering or solderless kits that can be

populated with a range of sensors might spur the interest of

people outside technically-minded DIY communities.

Alternatively, sensors with complex underlying circuits

might support easy alteration of device appearance, display,

or form factor. A parallel body of research can explore

intuitive instruction methods, including images, video, or

in-person demonstrations. More broadly, these approaches

can serve to playfully engage everyday citizens in designing

and building ubiquitous technologies that monitor issues of

personal concern.

Playful media

In addition, our light-hearted and playful medium

(balloons) led participants to explore and discuss air quality

throughout the city. During our public installations,

observers began to think about air quality in their

neighborhood (“diesel exhaust—that’s very bad thing”). In

our deployment with six everyday citizens, participants

visited a park and downtown area, walked through a nearby

neighborhood, and tethered the balloon to a bicycle,

describing their experiences as ‘fun’, ‘interesting’, or

‘surprising’. Moreover, the balloons inspired participants to

question and propose theories about air quality in their

surroundings: P1 and P2 wondered whether the gas from

their stove was emitting diesel exhaust; P3 hypothesized

that VOC levels were higher due to poor ventilation or a

nearby hospital; while W3 wanted to “see how the colors

would change” as he walked away from traffic.

Whereas current applications tend to rely on participants’

interest in the environment, community values, or monetary

compensation as motivations for sensing [e.g., 15], an

alternative body of work can encourage grassroots data

collection through tangible and playful media. Future

systems can explore incorporating play into the experience

of environmental sensing. Balloons, for instance, can be

used in range of new applications: balloons tethered to a

map showing data for a region; balloons that visualize other

environmental factors (e.g., water quality); or large-scale

projections that respond to onlookers’ balloons.

Moreover, other playful objects can be appropriated as

sensors to support explorations of overlooked or hard-toreach

spaces (similar to Wallbots [31]). For instance, air

quality sensors embedded in kites can encourage

engagement with and monitoring of air quality while

playing on a beach; remote-controlled boats, outfitted with

water quality sensors might inspire fun investigations of

local water sources; or digging toys that include soil sensors

could support interactive play with soil. Future applications

may also leverage familiar platforms (e.g., mobile phones)

to distribute games and contests that motivate playful

engagement with and reflection on the environment.

The power of spectacle

Air quality balloons enabled us and our participants to

create spectacles. Our installations in public locations have

attracted onlookers to approach the exhibit, ask questions,

and voice their opinions. Participants who received working

balloons actively sought attention by bringing balloons to

busier streets and inviting strangers to photograph and

discuss the project. Moreover, our workshop participants

who assembled their own balloons, tried to bring them

“where people would be”, sharing the project with friends,

family and strangers (through Flickr). The balloons served

as boundary objects, engaging people from different

communities, backgrounds, and geographic locations in

environmental discourse: one observer recommended

recording “changes over a long period of time” to P5,

another passerby told P1 about ‘the levels in his hometown’;

W4 used the balloon to illustrate the “concept of air

pollution” with his daughter, while W2 and his friends

discussed having balloons in “several locations thoroughout

Bangalore”; passersby who saw our public installations

suggested having “more people with balloons”, requesting

to take balloons to “spread the word”.

Our unconventional, vibrant, and public balloons served as

entry points for environmental discourse among a diverse

range of stakeholders. The installations and participants’

experiences with balloons encouraged speculations about

air quality between parents and children, local and

international groups of friends, as well as complete

strangers, either in person or by commenting on online

repositories of images. These findings suggest opportunities

to leverage spectacle computing for supporting new

interactions between people, technology, and space.

For instance, people with vested interests in a location (e.g.,

homeowners) might prefer longer-term environmental

spectacles in their neighborhood to facilitate local

discourse, or “community togetherness” [15]. Alternatively,

individuals who traverse different parts of the city (runners

or postal workers) might benefit from modular platforms

(e.g., Feral Robotic Dogs [25]), which support spectacles

and discussions across locations. Moreover, drawing from

our workshop participants’ willingness to modify the

balloon kit (with data logging, or different form factors

such as plastic globes), future systems can allow DIY

modification and re-appropriation of the spectacle. Finally,

the limitations of spectacles remain to be explored. When

(if ever) do glowing balloons and other similar installations

cease to be enticing How can the effectiveness of

spectacles be extended to inspire environmental discourse

over longer periods of time

Tangible media as an instrument of change

Emerging HCI research continues to explore the methods

and processes of activist groups [e.g., 21] in order to inform

the design of technologies for political change. We

contribute to this work by introducing a technology that can

be assembled by non-experts to make information about air

quality not only measurable and visible, but also

compelling and hard to ignore. Although none of our

participants were activists per se, they saw opportunities to

use air quality balloons to catalyze change. From wanting to

show the balloons to an “eco-friendly neighborhood”,

“students”, and “drivers”, to discussing the project with

friends abroad or the general public at large, participants

hoped to project collected data to other stakeholders who

might be willing to take action.

Our findings suggest opportunities to use tangible media as

a support tool for grassroots movements. On one hand,

other new and unexpected media may effectively entice and

empower bottom-up data collection: paper airplanes,

remote-control cars, dog collars, baby strollers, tree leaves,

or roller skates, to name a few, might serve as novel

vehicles for environmental sensing (similar to street

sweepers or pigeons [3, 12]). Data collected by such

distributed means might be aggregated into more traditional

representations (e.g., maps, graphs) to serve as political


Alternatively, tangible media might enable users to project

data within and across urban spaces in more playful and

unconventional ways. Wireless communication between

visualizations (e.g., balloons) and sensors can support

spectacles that are physically removed from the locations

being sensed. Users could place vibrant nodes in remote

spaces to draw the attention of activist communities, policy

makers or the general public to local issues.


We explored spectacle computing as an approach for

vibrantly projecting data into the public sphere and

inspiring environmental discourse. In two public

installations and a deployment with six everyday citizens

we evaluated a novel technology—large balloons that glow,

based on surrounding air quality. In addition, our kit

enabled six members of a local DIY community to

assemble their own, fully functional balloons. We leverage

our findings to propose play and spectacle as essential

elements for inspiring public discussion within and across

communities of stakeholders. We hope that our work

inspires other spectacle computing projects that engage

passive onlookers, routine visitors, and avid activists in

political and environmental discourse.


We thank all our participants for their feedback. Financial

support for this research was provided in part by the Heinz

Endowment. We thank our reviewers for their thoughtful

feedback, which will continue to shape our future research.


1. Aerophile, S.A. Air de Paris balloon.

2. Amphibious Architecture by The Living Architecture Lab

at Columbia University Graduate School of Architecture,

September 17 – November 7, 2009.

3. Aoki, P. M., Honicky, R. J., Mainwaring, A., Myers, C.

Paulos, E. Subramanian, S. Woodruff, A. A vehicle for

research: using street sweepers to explore the landscape of

environmental community action. CHI 2009, p. 375-384.

4. American Lung Association. City Rankings 2010: Most

Polluted Cities.

5. Buechley, L., Eisenberg, M. (2009) Fabric PCBs,

Electronic Sequins, and Socket Buttons: Techniques for E-

textile Craft. Journal of Personal and Ubiquitous

Computing, 13(2), p. 133-150.

6. Buechley, L., Eisenberg, M. (2008). The LilyPad Arduino:

Toward Wearable Engineering for Everyone. Wearable

Computing Column in IEEE Pervasive, v.7 n.2, p.12-15.

7. Buechley, L., Rosner, D., Paulos, E., and Williams, A.

(2009). Workshop: DIY for CHI: methods, communities,

and values of reuse and customization. In Proceedings of

the 27th international conference extended abstracts on

Human factors in computing systems (CHI EA '09). ACM,

New York, NY, USA, 4823-4826.

8. Burke, J., Estrin, D., Hansen, M., Parker, A., Ramanathan,

N., Reddy, S., Srivastava, M.B. Participatory Sensing. In

World Sensor Web Workshop. ACM, 2006.

9. Coalition for Clean Air. Pollutants & Health Effects.

10. Brynskov, M. Aarhus by Light. 2010.

11. Da Costa, B., Schulte, J., and Singer, B. AIR – Area’s

Immediate Reading.

12. Da Costa, B., Hazegh, C., Ponto, K. Pigeon blog.

13. Dearman, D., Truong, K. N. 2009. BlueTone: a framework

for interacting with public displays using dual-tone multifrequency

through bluetooth. In Ubicomp '09, p. 97-100.

14. Debord, G. The Society of the Spectacle. Zone, 1994.

15. DiSalvo, C. Marti, L., Coupland, J., Steiner, M. A 2009.

Local Issues, Local Uses: Tools for robotics and sensing in

community contexts. Creativity & Cognition, p. 245-254.

16. Eisenman, S. B., Miluzzo, E., Lane, N. D., Peterson, R. A.,

Ahn, G., and Campbell, A. T. 2009. BikeNet: A mobile

sensing system for cyclist experience mapping. In

Proceedings of SenSys '07. p. 87-101.

17. Figaro Gas Sensors.

18. Graffiti Research Lab. Throwies.

19. Grassroots Mapping. Balloon Aerial Photography.

20. Haque, U. Burble, London, UK 2007,

21. Hirsch, T., Henry, J. 2005. TXTmob: text messaging for

protest swarms. In CHI '05 extended abstracts on Human

factors in computing systems (CHI EA '05), p. 1455-1458.

22. Holmquist, L. E.. 2006. Sketching in hardware. Interactions

13, 1 (January 2006), p. 47-60.

23. Honicky, R., Brewer, E. A., Paulos, E., and White, R.

2008. N-smarts: networked suite of mobile atmospheric

real-time sensors. ACM SIGCOMM Workshop on

Networked Systems For Developing Regions, 2008.

24. Hooker, B., Gaver, W.W., Steed, A., & Bowers, J. The

Pollution e-Sign. Workshop on Ubiquitous Sustainability.

Ubicomp, 2007.

25. Jeremijenko, N., Arnold, J. J., Kavesh, W. Feral Robotic

Dogs. 2009.

26. Kaye, J., Williams, A., Oehlberg, L. Tinkering, Crafts and

Inventive Leisure Practices. In proceedings of the 2011

conference on Computer Supported Cooperative Work.

27. Kaprow, A. Assemblages, Environments & Happenings,

Harry N. Abrams, Inc.Publishers, New York, 1966.

28. Kim, S. and Paulos, E. 2009. inAir: measuring and

visualizing indoor air quality. Ubicomp '09, p. 81-84.

29. Kim, S., Paulos, E., and Gross, M. D. 2010. WearAir:

expressive t-shirts for air quality sensing. In Proceedings of

the fourth international conference on Tangible, embedded,

and embodied interaction (TEI '10), p. 295-296

30. Kuznetsov, S. Paulos, E.. 2010. Rise of the expert amateur:

DIY projects, communities, and cultures. In Proc. of the 6th

Nordic Conference on Human-Computer Interaction:

Extending Boundaries (NordiCHI '10), p. 295-304

31. Kuznetsov, S., Paulos, E, Gross, M.D. WallBots:

Interactive Wall-Crawling Robots In the Hands of Public

Artists and Political Activists. In Proc. of the 8th ACM

Conference on Designing Interactive Systems (DIS '10).

ACM, New York, NY, USA, 208-217.

32. Lamorisse, A. The Red Balloon.

33. Marx, L. 1867. The Fetishism of Commodities and the

Secret thereof. Das Kapital (1867).

34. NASA. New NASA Balloon Successfully Flight-Tested

Over Antarctica, December 2009.

35. Nena. 99 Luftballons. Released in 1983.

36. NOAA National Weather Service. Radiosonde


37. OFriel, K., CO2RSET.


39. Paulos, E., Honicky, R.J., Hooker, B. Citizen Science:

Enabling Participatory Urbanism. Handbook of Research

on Urban Informatics, 2008.

40. Pete's Sonic Art Research. Wire and Balloon Installation.

41. Piene, O. More Sky. MIT Press, 1973.

42. Rosner, D., Bean, J. 2009. Learning from IKEA hacking:

I'm not one to decoupage a tabletop and call it a day. In

Proceedings of the 27th international conference on Human

factors in computing systems (CHI '09), p. 419-422.

43. Seitinger, S., Taub, D. M., Taylor, A. S. 2010. Light

bodies: exploring interactions with responsive lights. In

Proc. of the fourth international conference on Tangible,

embedded, and embodied interaction (TEI '10), 113-120.

44. U.S. EPA. Health Assessment Document for Diesel Engine

Exhaust. U.S. Environmental Protection Agency, 2002

45. U.S. EPA.Volatile Organic Compounds.

More magazines by this user
Similar magazines